The outcome
Publish a clear synthetic performance without misrepresenting a real person or losing control of their likeness and voice.
Step by step
A workflow you can repeat.
- 01
Document the speaker or rights holder's consent, allowed script, channels, duration, audience, territories, revocation path, and synthetic-media disclosure.
- 02
Prepare a front-facing authorized image and clean authorized audio, removing metadata and private background details that are not needed.
- 03
Generate a short low-resolution test with the chosen avatar model, aspect ratio, neutral motion prompt, and no public sharing.
- 04
Review pronunciation, lip-sync, identity, gaze, teeth, hands, gesture, emotion, captions, background, and whether the performance implies unapproved claims.
- 05
Correct the source or prompt, render only the approved take, add visible disclosure and accessibility, obtain final sign-off, and delete unused identity assets where appropriate.
Working standard
What good use looks like.
- Get specific written consent.
- Test with a short clip first.
- Keep disclosure attached to the media.
Official references